Abstract
The problem of nonlinear alignment of functions is both fundamental and extremely important in pattern recognition. Most common approaches for alignment, including dynamic time warping (DTW), use penalized-\(\mathbb {L}^2\) minimization that has significant shortcomings, including asymmetry. A recent mathematical framework, based on an elastic Riemannian metric and square-root velocity functions, overcomes these shortcomings. The time warping problem is currently solved using a dynamic programming algorithm (DPA) which relies heavily on dense sampling of functions and can be computationally expensive. Here we present a novel theory (and algorithms) for finding the exact pairwise alignment between functions, which uses an efficient sampling of functions, restricted to their change points. In many cases, the computational cost for matching is reduced by orders of magnitude. We demonstrate the superiority of this method over the DPA using several simulated and real datasets.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Beg, M.F., Miller, M.I., Trouve, A., Younes, L.: Computing large deformation metric mappings via geodesic flows of diffeomorphisms. Int. J. Comput. Vis. 61(2), 139–157 (2005)
Bertsekas, D.P.: Dynamic Programming and Optimal Control. Athena Scientific, Belmont (1995)
Huang, W., Gallivan, K.A., Srivastava, A., Absil, P.A.: Riemannian optimization for elastic shape analysis. In: The 21st International Symposium on Mathematical Theory of Networks and Systems (MTNS 2014) (2014)
Keogh, E., Ratanamahatana, C.A.: Exact indexing of dynamic time warping. Knowl. Inf. Syst. 7, 358–386 (2005)
Kurtek, S., Wu, W., Christensen, G.E., Srivastava, A.: Segmentation, alignment and statistical analysis of biosignals with application to disease classification. J. Appl. Stat. 40(6), 1270–1288 (2013)
Lahiri, S., Robinson, D., Klassen, E.: Precise matching of PL curves in \(\mathbb{R}^N\) in the square root velocity framework (2015)
Liu, X., Muller, H.G.: Functional convex averaging and synchronization for time-warped random curves. J. Am. Stat. Assoc. 99, 687–699 (2004)
Rabiner, L., Juang, B.H.: Fundamentals of Speech Recognition. Prentice-Hall Inc., Upper Saddle River (1993)
Ramsay, J.O., Li, X.: Curve registration. J. Roy. Stat. Soc. Ser. B 60, 351–363 (1998)
Ramsay, J.O., Silverman, B.W.: Functional Data Analysis. Springer Series in Statistics, 2nd edn. Springer, New York (2005)
Sakoe, H., Chiba, S.: Dynamic programming algorithm optimization for spoken word recognition. IEEE Trans. Acoust. Speech Signal Process. 26(1), 43–49 (1978)
Sebastian, T.B., Klein, P.N., Kimia, B.B.: On aligning curves. IEEE Trans. Pattern Anal. Mach. Intell. 25(1), 116–125 (2003)
Srivastava, A., Klassen, E.: Functional and Shape Data Analysis. Springer Series in Statistics. Springer, New York (2016). https://books.google.com/books?id=0cMwDQAAQBAJ
Srivastava, A., Klassen, E., Joshi, S.H., Jermyn, I.H.: Shape analysis of elastic curves in Euclidean spaces. IEEE Trans. Pattern Anal. Mach. Intell. 33(7), 1415–1428 (2011)
Srivastava, A., Wu, W., Kurtek, S., Klassen, E., Marron, J.S.: Registration of functional data using Fisher-Rao metric (2011). arXiv:1103.3817v2
Trouvé, A., Younes, L.: Diffeomorphic matching problems in one dimension: designing and minimizing matching functionals. In: Vernon, D. (ed.) ECCV 2000. LNCS, vol. 1842, pp. 573–587. Springer, Heidelberg (2000). doi:10.1007/3-540-45054-8_37
Veeraraghavan, A., Chellappa, R., Roy-Chowdhury, A.: The function space of an activity. In: IEEE Conference on Computer Vision and Pattern Recognition, vol. 1, pp. 959–968 (2006)
Veeraraghavan, A., Srivastava, A., Chowdhury, A.K.R., Chellappa, R.: Rate-invariant recognition of humans and their activities. IEEE Trans. Image Process. 18(6), 1326–1339 (2009)
Zhou, F., Torre, F.: Canonical time warping for alignment of human behavior. In: Advances in Neural Information Processing Systems (2009)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
1 Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
Copyright information
© 2017 Springer International Publishing AG
About this paper
Cite this paper
Robinson, D., Duncan, A., Srivastava, A., Klassen, E. (2017). Exact Function Alignment Under Elastic Riemannian Metric. In: Cardoso, M., et al. Graphs in Biomedical Image Analysis, Computational Anatomy and Imaging Genetics. GRAIL MICGen MFCA 2017 2017 2017. Lecture Notes in Computer Science(), vol 10551. Springer, Cham. https://doi.org/10.1007/978-3-319-67675-3_13
Download citation
DOI: https://doi.org/10.1007/978-3-319-67675-3_13
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-67674-6
Online ISBN: 978-3-319-67675-3
eBook Packages: Computer ScienceComputer Science (R0)